AI Engineer - Product development

Role Overview

We are looking for an AI Engineer in Mumbai with 3-5 years of relevant experience who can design, build, and deploy a secure, offline, LLM-powered platforms . You will work on everything from model integration and document intelligence pipelines to backend services and system architecture, in close collaboration with product, UI, and client stakeholders.

This role requires strong hands-on engineering skills, deep understanding of NLP/LLMs, and experience working in security-constrained enterprise environments in the BFSI sector.

The ideal candidate has experience taking AI solutions from idea -> implementation -> deployment -> iteration and is comfortable treating AI capabilities as real world product features (not standalone experiments). 

Responsibilities

AI & LLM Engineering

• Deploy and fine-tune LLMs on on-premise / private infrastructure (no external API dependency)
• Build backend architecture by developing NLP pipelines and implement RAG (Retrieval Augmented Generation) pipelines using local vector databases
• Optimize models for performance, memory usage, and inference latency on internal hardware
• Design AI components with clear product use cases and user workflows in mind
• Translate functional requirements into AI capabilities that can be embedded into applications

Document Intelligence

• Design and develop secure backend services (Python-based preferred) to orchestrate: Document ingestion, LLM inference, Scoring and comparison logic
• Build robust pipelines to process multi-format documents (DOCX, PDF, scanned documents, etc.)
• Handle document chunking, embeddings, metadata tagging, and version comparison
• Design explainable outputs for document reviewers (traceability to source clauses)
• Integrate with Microsoft ecosystem : SharePoint document repositories and MS Office file formats
• Ensure the entire system functions fully offline within a restricted network

·      Apply cloud computing fundamentals (compute, storage, networking) effectively

Security & Compliance

• Follow enterprise-grade security practices: No external data transfer, secure credential and access handling, Role-based access control (RBAC) support
• Align implementation with client IT/security requirements
• Design systems with logging, monitoring, audits and traceability in mind
• Familiarity with confidential computing and private networking patterns is a plus (e.g., Bastion access, Private Link/Private Endpoints, private DNS, Key Vault/secret management).
• Ensure compliance with ethical AI practices and regulatory frameworks.

Collaboration & Ownership

• Collaborate with frontend engineers to support UI requirements
• Participate in solution design discussions with client stakeholders
• Own components end-to-end—from POC to production deployment
• Contribute to technical documentation of assumptions, risks, decisions and deployment runbooks
• Act as a technical interface with client IT and legal stakeholders to clarify requirements and acceptance criteria.
• Own end-to-end delivery of assigned AI features/components (scope, milestones, acceptance criteria).
• Support demos, walkthroughs, UAT readiness, and handover with documentation/runbooks.

Qualifications

Core Technical Skills

• 3–5 years of hands-on experience in AI / ML engineering and Strong proficiency in Python
• Strong grasp of computer architecture, data structures, system software, and machine learning fundamentals.
• Solid understanding of NLP fundamentals, transformer architectures, embeddings and semantic search
• Strong experience working with structured and unstructured data, including preprocessing, validation, and transformation for AI pipelines
• Hands-on experience with open-source LLMs (e.g., LLaMA, Mistral, Falcon, etc.)
• Experience deploying models locally or on private servers (not just cloud APIs)
• Experience with frameworks such as Hugging Face, LangChain / LlamaIndex (or similar orchestration frameworks)
• Vector databases (FAISS, Chroma, Milvus, etc.)
• Experience with prompt engineering and structured outputs
• Ability to plan work, estimate, and deliver independently in a client-facing or delivery driven environment.

Systems & Infrastructure

• Experience building and deploying backend services (FastAPI, Flask, or similar) in banking, finance or other regulated
• Familiarity with Linux environments and GPU-based inference
• Experience with working on containerization and clustering (Docker preferred)
• Experience working in restricted / air-gapped environments is a big plus
• Familiarity with logging, monitoring, and troubleshooting of deployed services

·      Experience with information security and secure development best practices.

Education

• Bachelor’s or Master’s degree in: Computer Science, Data Science or Artificial Intelligence
• Exposure to banking, finance, or regulated enterprise environments
• Experience optimizing models for low-latency inference
• Familiarity with UI-driven AI workflows (human-in-the-loop systems)